ChatGPT Gov
Last reviewed
Jun 3, 2026
Sources
9 citations
Review status
Source-backed
Revision
v1 · 1,548 words
Improve this article
Add missing citations, update stale details, or suggest a clearer explanation.
Suggest edit
CtrlK
Last reviewed
Jun 3, 2026
Sources
9 citations
Review status
Source-backed
Revision
v1 · 1,548 words
Add missing citations, update stale details, or suggest a clearer explanation.
ChatGPT Gov is a version of OpenAI's ChatGPT tailored for United States government agencies, first announced on January 28, 2025. Built on the ChatGPT Enterprise product, it lets agencies deploy the service inside their own Microsoft Azure commercial cloud or Microsoft Azure Government environment so they can manage their own security, privacy, and compliance requirements when handling non-public, sensitive data. In June 2025, OpenAI folded ChatGPT Gov and its other public-sector work into a broader program called OpenAI for Government, which it launched alongside a roughly $200 million pilot contract with the U.S. Department of Defense.
ChatGPT Gov was introduced as an additional way for U.S. federal, state, and local agencies to access OpenAI's frontier models within infrastructure they control.[1] OpenAI presented the product as a response to demand it said it was already seeing across government: the company reported that, since the start of 2024, more than 90,000 users across more than 3,500 U.S. federal, state, and local government agencies had sent over 18 million messages using ChatGPT to support their work.[1][2]
OpenAI has consistently distinguished three related things. ChatGPT Gov is the self-hosted product for government. ChatGPT Enterprise is OpenAI's fully managed commercial offering, on which ChatGPT Gov is based and which OpenAI separately worked to bring into government compliance regimes. OpenAI for Government, announced later, is an umbrella program that consolidates these products together with the company's public-sector partnerships and contracts.[3]
Kevin Weil, OpenAI's chief product officer, framed the launch around efficiency, saying that "by adopting AI, government agencies can work smarter, faster and more creatively."[2]
The defining characteristic of ChatGPT Gov is the deployment model. Rather than using OpenAI's hosted service, agencies run the product within their own Microsoft Azure commercial cloud or Azure Government tenancy, on top of the Microsoft Azure OpenAI Service.[1] OpenAI argued that self-hosting in this way makes it easier for agencies to satisfy stringent cybersecurity and data-handling frameworks, citing IL5 (Impact Level 5), CJIS (Criminal Justice Information Services), ITAR (International Traffic in Arms Regulations), and FedRAMP High among the relevant standards.[1] Because the workloads run in environments the agency already controls, ChatGPT Gov is positioned for non-public but sensitive information, with OpenAI pointing to defense, intelligence, and healthcare use cases.[2]
Functionally, ChatGPT Gov mirrors much of ChatGPT Enterprise. OpenAI said it includes the ability to save and share conversations inside a government workspace, upload text and image files, and use the model to analyze and summarize complex documents.[1][2] At launch, OpenAI also said it would continue pursuing FedRAMP Moderate and FedRAMP High accreditation for its fully managed ChatGPT Enterprise product, and that it was evaluating expanding ChatGPT Gov into Microsoft Azure's classified regions.[1]
| Aspect | ChatGPT Gov detail |
|---|---|
| Announced | January 28, 2025 |
| Based on | ChatGPT Enterprise |
| Hosting | Agency's own Microsoft Azure commercial cloud or Azure Government, via the Azure OpenAI Service |
| Data scope | Non-public, sensitive (not classified) data |
| Cited frameworks | IL5, CJIS, ITAR, FedRAMP High |
| Sample features | Save/share conversations in workspace, upload text and image files, document summarization |
| Stated roadmap | FedRAMP Moderate and High for ChatGPT Enterprise; evaluating Azure classified regions |
On June 16, 2025, with public announcement the following day, OpenAI launched OpenAI for Government, an initiative that consolidates the company's work with the public sector under a single program.[3][4] OpenAI described it as bringing together previously announced customers and partnerships, along with ChatGPT Gov, as the company expands government work.[3] The program offers federal, state, and local agencies access to OpenAI's models within secure and compliant environments, including through ChatGPT Enterprise and ChatGPT Gov, and, on a limited basis, custom models for national security purposes.[3][4]
The centerpiece of the launch was a Department of Defense pilot. The DoD's Chief Digital and Artificial Intelligence Office (CDAO) awarded a contract worth up to $200 million to OpenAI Public Sector LLC.[4][5] The agreement was structured as an Other Transaction Agreement (OTA), a flexible Pentagon contracting mechanism used to speed prototype and research work with nontraditional vendors.[5] The stated purpose is to "develop prototype frontier AI" addressing national security challenges across both back-office enterprise functions and frontline operations, including analyzing acquisition and program data, supporting proactive cyber defense, improving how service members and their families access health care, and prototyping agentic workflows, meaning semi-autonomous AI agents that can carry out routine tasks.[4][5]
At the time of award, about $1.9 million was obligated to OpenAI from fiscal year 2025 research, development, test, and evaluation funds.[5] The work is to be performed primarily in the National Capital Region, with an estimated completion date of July 2026.[5] OpenAI emphasized that all use cases under the program must comply with the company's usage policies, which prohibit using its tools to develop or use weapons.[4]
| Date | Event |
|---|---|
| January 28, 2025 | OpenAI announces ChatGPT Gov |
| January 30, 2025 | OpenAI announces expanded work with U.S. National Laboratories |
| June 16, 2025 | DoD CDAO awards up to $200 million OTA to OpenAI Public Sector LLC |
| June 17, 2025 | OpenAI publicly launches OpenAI for Government |
| August 6, 2025 | GSA and OpenAI announce a OneGov deal offering ChatGPT Enterprise to federal agencies for $1 per agency for one year |
| July 2026 | Estimated completion of the DoD prototype contract |
OpenAI for Government gathered a range of existing public-sector relationships under one banner. Named partners and customers include the Air Force Research Laboratory, U.S. Department of Energy National Laboratories (including Los Alamos), NASA, the National Institutes of Health (NIH), the Treasury Department, the Department of Homeland Security (DHS), and the Commonwealth of Pennsylvania.[3][4]
Several of these predate the umbrella program. In January 2025, OpenAI announced an expanded partnership with the U.S. National Laboratories under which its models would be deployed to accelerate scientific research and bolster national security readiness, including work at Los Alamos, Lawrence Livermore, and Sandia.[6] As part of that effort, OpenAI said it would work with Microsoft to deploy an o-series reasoning model on Venado, an NVIDIA-based supercomputer at Los Alamos National Laboratory, as a shared resource for researchers across the three labs.[6] The Air Force Research Laboratory had adopted ChatGPT Enterprise for administrative use cases such as improving access to internal resources, basic coding, and AI education.[1] At the state level, OpenAI reported that employees in a Commonwealth of Pennsylvania pilot found ChatGPT reduced time spent on routine tasks by about 105 minutes per day.[3]
Government adoption broadened further after the program's launch. On August 6, 2025, the U.S. General Services Administration (GSA) and OpenAI announced a partnership under GSA's OneGov strategy to make ChatGPT Enterprise available to participating federal agencies for a nominal fee of $1 per agency for one year, followed by a 60-day period of unlimited access to advanced models at no additional cost.[7] Press coverage noted that the arrangement supplanted previously contracted pricing reported at roughly $191 per user per year.[8] The GSA deal came shortly after ChatGPT, Google's Gemini, and Anthropic's Claude were added to GSA's Multiple Award Schedule, and Anthropic indicated it was pursuing a similar $1 arrangement.[8]
ChatGPT Gov and OpenAI for Government were part of a wider 2025 push by major AI developers to win U.S. public-sector and defense business. The same Pentagon effort that produced OpenAI's award was reported to extend to competitors: in mid-2025, the CDAO announced frontier-AI contract actions involving Google, Anthropic, and Elon Musk's xAI in addition to OpenAI, each potentially worth up to $200 million.[9] On the procurement side, the near-simultaneous GSA Multiple Award Schedule additions and $1 OneGov offers from multiple vendors signaled aggressive competition to become the default AI tool inside federal agencies.[7][8]
OpenAI's government strategy also intersects with its broader infrastructure ambitions, including the Stargate Project data-center buildout announced in January 2025, and with Azure OpenAI Service, the Microsoft platform on which ChatGPT Gov runs. Taken together, the 2025 announcements positioned OpenAI to serve government customers across three layers: a self-managed product for sensitive data (ChatGPT Gov), broad low-cost access for the general federal workforce (ChatGPT Enterprise via GSA), and bespoke national-security work (custom models and the DoD prototype contract), all coordinated through the OpenAI for Government program.